AI

Recommendation is one of the most economically important applied ML problems, powering the ranked feeds, 'you might also like' rails, and autoplay queues that shape what billions of people watch, buy, and read. This course covers the two classic recommendation families and how they differ, the cold-start problem that breaks both of them in different ways, matrix factorization as the technique underlying much of modern collaborative filtering, why implicit feedback dominates real systems despite being noisier than explicit ratings, and the evaluation traps that let a recommender look great on paper while quietly narrowing what people see or failing to move the metric it was built for.