This course is divided into three parts. The first section will be a theoretical introduction to recommender systems. It will consist of a quick maths refresher. Students will be equipped to understand the basics underlying modern recommender systems, from matrix decomposition-based methods to deep learning.
The second section will be a practical notebook session. Students will interact with a set of Jupyter Notebooks, allowing them to get hands-on experience with how recommender systems are trained and put in production.
The third section will be a class discussion over policy issues regarding recommender systems, privacy and data collection. This discussion will be fuelled by the theoretical and practical coding sessions done earlier during the course.
