Learn
Learn by doing.
Interactive courses and simulations on AI and data, built to make hard ideas feel obvious.
New: machine learning simulations
Simulation library
Learn AI by touching it.
Every simulation is a small experiment: do something, watch the algorithm react and complete three challenges. Your progress is saved on this device.
Machine learning
Linear regression
Find the straight line that best fits your data, and see why one odd point can pull it off course.
Gradient descent
Watch a model learn by walking downhill on its own error, one step at a time.
Logistic regression
Turn a straight line into a probability, and discover what a straight line cannot separate.
Support vector machine
Find the widest street between two groups, and see that only a few points on its edges really matter.
k-nearest neighbours
Classify a new point by asking its closest neighbours to vote.
Naive Bayes
Classify with probabilities and a bold shortcut: within each class, assume every feature is independent of the others.
Decision tree
Split the data with simple yes or no questions, and see how a deeper tree starts memorising noise.
Random forest
Grow many slightly different decision trees and let them vote. The crowd is wiser than any single tree.
Gradient boosting
Build a strong model from many weak ones, each tree fixing what the previous ones still get wrong.
Overfitting
Make a curve more and more flexible and find the point where it stops learning and starts memorising.
Regularisation (ridge and lasso)
Tame an overfitting model by charging it for large weights, and see lasso switch the weakest features off.
Cross-validation
Test a model fairly by letting every part of the data take a turn as the test set.
k-means clustering
Find groups in data that nobody has labelled, two simple steps at a time.
Hierarchical clustering
Join the two closest groups again and again, and grow a family tree of your data.
DBSCAN clustering
Find clusters of any shape by following crowded regions, and leave lonely points out as noise.
Anomaly detection
Find the odd ones out by cutting the data at random: lonely points are isolated in just a few cuts.
Principal component analysis
Find the direction in which your data varies the most, and squeeze two dimensions into one.
Deep learning
Perceptron
Meet the first learning machine: one artificial neuron that corrects its line one mistake at a time.
Neural network
Stack neurons in layers and watch a network learn shapes that no straight line could ever draw.
Activation functions
Compare the small curves that let neural networks bend, and see why some of them make deep networks hard to train.
Backpropagation
Follow the error backwards through a tiny network and see how the chain rule tells every weight which way to move.
Convolutional network
Slide a small filter across a picture and watch it find edges: the trick that lets networks see.
Recurrent network
Give a network a memory that loops back on itself, and see how quickly that memory fades.
Attention
See how a transformer decides which words to look at, with nothing more than dot products and a softmax.
Generative AI
Tokens
Next-word prediction
Temperature and sampling
Embeddings
Retrieval (RAG)
What's coming
Understand it by playing with it.
Every lesson will be something you can touch: change a value, watch a model react and see why it works.
- Interactive courses
- Short, hands-on lessons on AI, machine learning and data, from first principles to real projects.
- Live simulations
- Change one thing and watch what happens, so you learn why, not only how.
- Certificates
- Planned: optional assessments that earn a Chromoworks certificate.