How a network computes, why gradients vanish, and what makes depth trainable
In Machine Learning you planted a straight fence through a scatter of points and read off which side each point landed on (ml.9). A neuron, also called a unit, is that same fence with one addition: after computing the score it passes that score through a function before reporting an answer. That is the entire object. Every network in this course is built by stacking copies of it.
🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.
▶ The Neuron: a Boundary That Learns