About this course
This course will provide an entry-level introduction to graph machine learning (GML), ones of the fastest-growing fields in AI and machine learning. GML aims at processing and analysing data that are collected in network domains, such as biological, social and financial networks, and hence are supported by graph structures. The course will first give a brief introduction on network science and graph representations, covering concepts such as node centrality, clustering coefficient, and community detection. It will then move onto a short introduction to graph signal processing, which builds fundamental tools to process signals on graphs, including frequency analysis of graph signals, graph Fourier transform, and filters and convolution. This will then enable discussion about graph machine learning models, in particular convolutional neural networks on graphs and message-passing neural networks. We will end the course with current developments in the field and real-world applications.
Learning Outcomes
By the end of the course, you will:
- Understand basic concepts in network science and graph representations.
- Understand basic concepts and tools in graph signal processing, including frequency-domain analysis, graph Fourier transform, filtering and convolution.
- Recognise basic forms of machine learning problems on graphs; understand basic design principles of graph neural networks (GNNs), including both convolutional and message-passing GNNs.
- Be able to apply the tools learned in the course to real-world network analysis.
Who is this course suitable for?
This course would suit STEM students with intermediate level experience in artificial intelligence, machine learning, and graph machine learning concepts and techniques, including those undertaking, or looking ahead to, graduate level study or research.
Course Convenor: Prof Xiaowen Dong
Prof Dong is an Associate Professor in the Department of Engineering Science, where he is an academic member of both the Machine Learning Research Group and the Oxford-Man Institute. He is also a Tutorial Fellow at Lady Margaret Hall. Prior to joining Oxford, he was a postdoctoral associate in the MIT Media Lab, where he remains as a research affiliate, and received his PhD degree from the Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland.
Dates and availability
Available as a Residential or Online course on the following dates:
Session 1: 28th June - 16th July 2027