Machine Learning Basics
How machine learning works at a conceptual level: show the computer many examples, it finds patterns, then it makes predictions about new things; hands-on experience with Teachable Machine or similar tool
Lesson
Set out your lesson
“Machine Learning Basics means: show many examples, the system finds patterns, then it predicts about something new. We will train a human “model” with cards, test a new card, talk about mistakes, then check a fresh prediction.”
Bring to the table
- Paper cards: many circles vs triangles (or sock vs spoon photos)
- Optional: adult-supervised Teachable Machine in a browser later—not required for the table lesson
Extra for the adult
Worked example. After many circle examples, a new round circle is predicted “circle.” A weird oval might be mislabeled—that is a model mistake.
Optional challenge. For Machine Learning Basics, deliberately train with biased examples (only red circles) and show how predictions fail on a blue circle. Optional: adult-led Teachable Machine image project another day.
What to notice. Keep one parent note: train examples; predictions; mistake talk; support used. Optional tool use is extra, not required evidence. This session samples the concept, not mastery of ML tools.
Share and support
Compare teaching approaches
Teaching approaches
Gakuva’s recommended starting approach is selected. You can still compare every option.
1 approach
Machine Learning Basics means: show many examples, the system finds patterns, then it predicts about something new. We will train a human “model” with cards, test a new card, talk about mistakes, then check a fresh prediction.
Spend about 14 minutes practicing example→pattern→predict with card sorting (optional later tool), then checking a new prediction.
Family records and proof options
Keep a record of your learning
Save photos, notes, and attendance. Export a proof packet for compliance or portfolio reviews.
Practice this concept
Start practiceMachine learning basics practice
Use examples to predict, and expect mistakes.
Free · no login · no ads · Practice answers and completion are not saved. Parent observation remains the lesson record.
Learning resources
Extra practice links — they do not replace Gakuva’s teaching approach.
Google Creative Lab · Practice link — not the full lesson
Teachable Machine · Image
Before you open · source notes
Hands-on machine-learning basics: gather examples, train, and test. Emphasize that the computer only learns from the examples you give it.
Free · no login · no ads observed · webcam optional
Before you openSave before/after screenshots (classes filled → prediction) and a parent note in the learner’s words: “It learned from my examples.”
Opening this link does not save learning or assess mastery.
Source, access and review notes
Unknown — link only
Code.org · Practice link — not the full lesson
AI for Oceans
Before you open · source notes
Guided story-mode introduction to training data and model testing for younger learners before open-ended Teachable Machine work.
Free · no login required to play · account optional to save · no ads observed
Before you openSave a progress screenshot and a parent note explaining training data vs a later test in kid language.
Opening this link does not save learning or assess mastery.
Source, access and review notes
Unknown — link only
Check understanding
- Describe the three steps of machine learning in simple terms (give examples, find patterns, make predictions)
- Train a simple model using a tool like Teachable Machine and describe what happened
- Explain why showing more and better examples makes the model more accurate
“Could your child describe how you teach a computer to recognise something — by showing it lots of examples until it spots the pattern?”
Curriculum record
- Type
- Conceptual
- Subject
- Computing
- Domain
- Artificial Intelligence
- Age range
- Ages 7–9