AI learning hub
Learn how AI works by building it yourself: 47 challenges across neural networks, machine learning, Transformers and reinforcement learning, with no AI libraries. Every challenge runs in your browser with instant tests and an AI review.
Arrays and matrices are the language of AI. Skip this step if you already know them.
Python for AI →Each module starts with a short lesson with runnable code that explains the idea in plain words.
All AI lessons →Implement the idea yourself in Python, C++ or C#, with instant tests and an AI review.
Neurons, backpropagation and optimizers
Regression, classification and clustering
Tokenizers, attention and text generation
Search, games and reinforcement learning
Multiply two vectors element by element and add the results: the core operation of every neuron.
Apply sigmoid, ReLU, Leaky ReLU or tanh to every value in a list.
Compute the output of one neuron: weighted sum of the inputs plus a bias, passed through sigmoid.
Compute the outputs of a fully connected layer of neurons with a choice of activation.
Run inputs through a hidden ReLU layer and a sigmoid output layer.
Measure how wrong a model's predictions are with mean squared error, mean absolute error or binary cross-entropy.