How Models Learn

Loss curve

The loss plotted against one parameter, such as the slope of a line, with everything else held fixed or chosen at its best.

What it can do cheaply is measure how steep the loss is right where it stands — the slope of the loss curve, not to be confused with the slope of the house line.
From Gradient descent, step “Lost in the fog”

See it in action

Try loss curve hands-on in the lesson Gradient descent, part of the AI fundamentals course.

Open the lesson