How Models Learn

Computational graph

A diagram of a calculation broken into simple steps, with arrows showing which values feed into which.

Written as a computational graph, every node does one simple step: u = w·x, then ŷ = u + b, then the error e = ŷ − y, then L = e².
From The chain rule, step “Break a model into a graph”

See it in action

Try computational graph hands-on in the lesson The chain rule, part of the The maths behind AI course.

Open the lesson