THE GENIUS PROJECT
Hub Week 1 Week 2 Week 3 Week 4 Final Quiz
Monday · Virtual · 10:00 AM

Machine Learning at Work: The Five Models

How machines learn from data, and which model fits which job: predicting a number, predicting a yes/no, finding groups, making rule-based decisions, and forecasting the future. Work through the five labs on this page before Friday.

Wednesday · In Person · 10:00 AM

Satellite AI Challenge

Explore how AI reads satellite and earth data, the same imagery used to watch storms form in the Atlantic and track sargassum heading to our beaches.

Friday · Virtual · 10:00 AM

Solving Your Parents' Problems with AI

Bring a real problem from home. Together we decide which of the five models fits it, and start building the solution.

Model 1 · Linear Regression The Sky Juice Stand 🧊

Linear regression predicts a number by drawing the best straight line through data. Here, each blue dot is one day at a Half Way Tree sky juice stand: how hot it was, and how many cups sold. Hotter day, more cups. The model's job is to find the line that fits that pattern best.

How to use this lab
  1. Look at the dots first. Can you see the upward pattern with your own eyes?
  2. Press Fit the line 📈. The gold line starts as a bad guess and improves in rounds. The thin coral lines show each day's error; watch them shrink.
  3. When the line settles, use the Tomorrow's temperature slider. The gold dot rides the line and reads off the prediction.
The flat gold line is an untrained model. Press "Fit the line".

Model 2 · Logistic Regression Will the Gully Flood? 🌊

Sometimes you do not want a number, you want a yes or no with a probability. Logistic regression fits an S-shaped curve that turns any input into a probability between 0% and 100%. Each dot below is a past storm day: rainfall along the bottom, and whether the gully flooded (top row) or not (bottom row).

How to use this lab
  1. Notice the messy middle: around 50 to 70 mm of rain, some days flooded and some did not. A plain yes/no rule fails there.
  2. Press Fit the S-curve and watch the model draw its probability curve through the mess.
  3. Move the Today's rainfall slider. The gold dot slides along the curve and the status line reads the flood probability.
  4. Find the rainfall where the probability crosses 50%. That is the model's decision boundary.
Press "Fit the S-curve" to train the model.

Model 3 · K-Means Clustering Who Comes to Doctor's Cave Beach? 🏖️

The first two models learned from labelled examples. K-means is different: it finds groups nobody labelled. Each dot is a beach visitor in Montego Bay: hours spent on the beach, and money spent. Nobody told the computer who is who. Watch it discover the groups itself: this is called unsupervised learning.

How to use this lab
  1. Press Drop 3 centroids. The three big diamonds are guesses at where the group centres might be.
  2. Press Next step. First press: every visitor is coloured by its nearest diamond. Second press: each diamond moves to the middle of its colour group.
  3. Keep pressing Next step. Assign, move, assign, move. Watch the diamonds settle.
  4. When the diamonds stop moving, the model has converged. Read the three groups it found in the status line.
Unlabelled data: the computer knows nothing about these visitors yet.

Model 4 · Decision Tree Should School Close for the Storm? 🌀

A decision tree is a model that learned which questions to ask, and in which order. Each split was chosen from data because it best separates the outcomes. You are the principal this morning: walk the tree and let it guide your call.

How to use this lab
  1. Read the highlighted question at the top of the tree.
  2. Press Yes or No below the diagram to answer it.
  3. Follow the glowing path down to the next question, and keep answering until you reach a decision box.
  4. Press Start over and try a different storm morning. Notice how one changed answer can change the whole path.
Question 1: is the wind above 60 mph?

Model 5 · Time Series Forecast Forecast the Tourist Season ✈️

A time series model learns from the past of one measurement to forecast its future. This chart shows two years of monthly visitor arrivals to Jamaica (in thousands). See the repeating shape? High season in winter, a summer bump, a September to October dip. That repeating shape is called seasonality, and it is exactly what the model learns.

How to use this lab
  1. Study the blue history line. Find the two winter peaks and the two autumn dips.
  2. Press Forecast the next 6 months. The gold dashed line is the model's forecast; the shaded band is its uncertainty, and it widens the further out you go.
  3. Press Reveal what actually happened and compare. Was the truth inside the band?
Two years of history. The next six months are unknown.

Mini Game Feature Picker: What Should the Model Study?

Whichever of the five models you choose, it only learns from the data you feed it. A "feature" is one input column. You are building a model to predict flooding in Portmore: choose wisely.

How to play
  1. Tap chips to select them. A selected chip turns teal. Tap again to unselect.
  2. Select exactly four features that genuinely help predict floods.
  3. Press Check my features. Green means useful signal, coral means noise.

📬 Week 2 Assignments (Mission Days)

1. Model Matchmaker (Tuesday). Write down five problems from your own life or your community, one for each model type. Example: "predict my phone data usage next month" is a time series forecast; "will the bus come before 7:15, yes or no" is logistic regression territory. Bring your five to Friday's session.

2. Problem Hunt (Thursday). Interview one parent or guardian. Ask them: "What is the most annoying, repetitive problem in your week?" Write it down in one sentence, then decide which of the five models could attack it.

3. Flood model rematch (weekend). Go back to the AI Quests flood mission and try to beat your Week 1 accuracy. You now know what the model is doing under the hood. Submit your new score at the same link.

Replay the Flood Prediction quest → Submit your improved score →