Back to the 2026 paper

Module 3: Linear Classification and Advanced Methods

20262m

During backpropagation, the main purpose of propagating the error backward is to (i) Generate new training data (ii) Normalize the input (iii) Remove hidden layers (iv) Compute gradients for updating network weights

Worked SolutionAI Assisted

Answer: (iv) Compute gradients for updating network weights

Backpropagation applies the chain rule from the output layer backward to calculate loss gradients with respect to weights and biases. An optimizer uses these gradients to update the parameters.

Similar questions

Student FinanceWeb Platform

Pocketly — Personal Finance for Students

Keep your hostel and college expenses under control. Track accounts, log spending in seconds, and stay within your monthly budget.