Deep Learning & AI in Sport

This eight-week online course is the most advanced programme in Twelve Football's data-science track, delivered in partnership with SkillCorner and aimed at experienced Python users, data scientists and sports analysts working in football, basketball or ice hockey. The curriculum applies cutting-edge AI methods to large tracking and performance datasets — covering deep learning, graph neural networks, autoencoders and transformers — and walks participants through Expected Possession Value, Pitch Control, spatiotemporal modelling, clustering and dimensionality reduction (PCA, t-SNE, UMAP), reinforcement learning for in-game action valuation, and training football language models. Coursework includes a group project building applications on tracking data and scout-report databases.

Delivered online by Dr. Pegah Rahimian — football data scientist at Twelve Football and post-doctoral researcher at Uppsala University — alongside Professor David Sumpter, with two lectures per week and hands-on coding sessions in Python using TensorFlow / PyTorch. Invited industry guest experts contribute throughout, and participants work in groups towards final project presentations. The next intake begins on 20 August 2026.