Grade 7 · AI Literacy
Explain that a neural network is layers of simple connected units that learn by adjusting the weights of their connections.
Most modern AI is built on an artificial neural network — a web of many tiny units ('neurons') loosely inspired by the brain. Each neuron takes in numbers, weights them, adds them up and passes on a simple signal. On their own the units are simple; connected in layers by the thousands or millions, they can recognise faces, translate languages or write text.
A network learns by adjusting the strength ('weight') of its connections. During training it compares its answer to the correct one, measures the error, and nudges the weights to make the error smaller — over and over, on huge amounts of data. Nobody writes a rule for each case; the learned pattern lives in the final set of weights.
Have a go right here. Nothing is saved and you can retry as much as you like. Create a free account when you want to save your progress and get Archie, your tutor.
Create a free account to save your progress, track mastery and ask Archie, your tutor, for a hand.
Already with us? Open this lesson in learncoach.ch for mastery tracking and Archie, your tutor.