How This Connects to What You Already Know

Machine learning is the middle layer of the AI/ML/deep-learning nesting: AI is the broadest goal (getting software to do things that normally need human intelligence), machine learning is one major approach to reaching it (learning patterns from data instead of hand-written rules), and deep learning is a specific kind of machine learning built on neural networks.

This track goes deeper specifically into machine learning as a field: the different types of problems it solves, how a model actually gets evaluated, and the classic techniques that came before — and still sit alongside — the deep learning systems behind modern chatbots and image generators.

Not every machine learning system today is a deep learning system. Plenty of real, everyday ML — fraud detection, recommendation systems, medical risk scoring — runs on the kind of classic techniques this track covers, precisely because they're often simpler, faster, and easier to understand than a deep neural network, while working just as well for the specific problem at hand.

How does machine learning fit into the broader AI/ML/deep-learning relationship?

Many real, everyday ML systems (like fraud detection) run on classic techniques rather than ____, because those techniques are often simpler and just as effective for the problem.

Every machine learning system in real-world use today is built on deep learning specifically.