Coding and Gaming Design - Intro to Machine Learning

Intro to Machine Learning (ML) demystifies one of the most important and least understood technologies of our time—through hands-on experience that requires zero prior coding knowledge. Using Google's Teachable Machine, students train a real image classification model from scratch and discover firsthand how machine learning actually works.

Students begin with a clear, jargon-free explanation of what machine learning is and how it differs from traditional programming. Rather than writing rules, ML systems learn patterns from examples—and this lesson makes that abstract idea concrete through immediate experimentation.

Students collect training data, feed it into Teachable Machine, and train a model to distinguish between two categories of their choosing—hand gestures, facial expressions, objects, or anything they can capture on camera. They test the model in real time, observe where it succeeds and where it fails, and investigate why.

The lesson digs into what those failure modes reveal: that ML models are only as good as their training data, that small datasets produce unreliable results, and that bias in training examples produces bias in outputs. Students experience these phenomena directly rather than just reading about them.

Students leave with a trained, working ML model they built themselves—and a foundational understanding of machine learning that demystifies the technology powering much of modern AI.

ISTE Standards 1 and 5 aligned.