Foundations advanced

Gradient Descent

The optimisation method that finds better parameters by repeatedly stepping downhill on the loss.

Gradient descent computes which direction each parameter should move to reduce loss, then takes a small step that way. Repeat millions of times and the model converges. Nearly all modern training uses a variant of it, usually Adam.

In practice: Like walking downhill in fog: you cannot see the valley, but you can feel the slope under your feet.

Where this comes up