Computer Science I / Multivariable Calculus and Gradients
Practice question · Fill in the blanks

Complete the update rule of gradient descent.

To minimise a function, each step moves from the current point ______.

Word bank: along the gradient by a learning-rate amount · opposite the gradient by a learning-rate amount · perpendicular to the gradient · to wherever the gradient is largest

Hints
  1. The gradient is the steepest ascent direction, and you want the opposite of ascent.
  2. The learning rate scales the step, it does not choose its direction.
Show the answer

To minimise a function, each step moves from the current point opposite the gradient by a learning-rate amount.

Why

Descent negates the gradient and scales it by the learning rate. Moving perpendicular to the gradient would trace a contour and change nothing, and heading for the largest gradient would climb the steepest slope rather than descend it.

Read the lesson: Multivariable Calculus and Gradients →

Practise Multivariable Calculus and Gradients

The app has 6 more questions on this lesson, and keeps your place in the course. Computer Science I is free to start.

More questions on Multivariable Calculus and Gradients