🤖 Train Your Own AI!

You are the teacher — show the AI examples, press Train, and watch it learn

›

Step 1

👆

Place Animals

Pick Pet or Wild, click the board

›

Step 2

⚡

Train the AI

Click "Train AI!" and it finds the pattern

Step 3

🎨

See the Zones

Colors show what the AI guesses anywhere

🐭 Small ← Size of Animal → Big 🐘

🎮 You're the Trainer!

Pick an animal type, then click anywhere on the board to place it.

What are you adding?

👀 Examples added for you! Add more, then click Train AI!

🐾 Pets: 3 🌿 Wild: 3

Accuracy: --

🧑‍🏫

You Are the Teacher

The AI knows nothing at the start. Every example you place teaches it something new. No examples = no learning!

🔍

AI Spots the Pattern

The AI finds the center of all Pet examples and all Wild examples, then decides which region belongs to which.

🔮

AI Makes Guesses

The colored zones show what the AI predicts for spots it has never seen. More spread-out examples = better guesses!

What is machine learning?

Machine learning is a way to teach computers by showing them examples — instead of writing every rule by hand. The more good examples you give, the smarter the AI gets. You just did exactly what real AI engineers do!

How did the AI decide the zones?

It found the center point of all your Pet examples and the center of all your Wild examples. For any new spot it asks: "Which center am I closer to?" That rule is called a centroid classifier — simple but effective!

Why does accuracy drop when animals are mixed?

When a Pet and a Wild animal are placed close together, the AI cannot easily tell them apart — just like a human might confuse a fox and a dog! Researchers call this the overlap problem, and solving it is a huge part of making AI smarter.

Key words

Training data — examples you show the AI to teach it Model — the pattern the AI learned from your examples Prediction — the AI's best guess for a new situation Accuracy — how often the AI's predictions are correct Centroid — the center point of a group of examples

🎯 Try this challenge

Can you get 100% accuracy? Place Pets in the bottom-left and Wild animals in the top-right, then train. Now add a pet into the wild zone area — does accuracy drop? That gap between easy and hard data is the 'overlap problem' real ML engineers deal with!

Continue Learning

AI for Students