Quiz Space

Deep Learning · Quiz 1 · 26 Oct 2025 · September 2025 term · Set 1

Question 5: A data scientist is training a deep learning model on a m…

Question 5

+3 marksOne correct option

A data scientist is training a deep learning model on a massive dataset containing billions of data points. They find that using traditional Batch Gradient Descent (GD) is computationally infeasible. What is the primary reason that makes Batch GD impractical for this task, thus requiring an alternative like Stochastic Gradient Descent (SGD)?

  1. A

    It is impossible to fit the entire large dataset into the computer’s memory(RAM) at once.

  2. B

    The gradient calculated by Batch GD is often too noisy and inaccurate, leadingto poor convergence.

  3. C

    Batch GD is guaranteed to get stuck in sharp local minima, whereas SGD canescape them.

  4. D

    Batch GD requires calculating the gradient of the loss function with respect tothe parameters over the entire dataset for a single update, which is computationally very expensive.

Show answer

Correct answer

  • D

    Batch GD requires calculating the gradient of the loss function with respect tothe parameters over the entire dataset for a single update, which is computationally very expensive.

Question 5 of 22 in the IIT Madras BS Deep Learning (Deep Learning) Quiz 1 paper sat on 26 Oct 2025, in the September 2025 term (IIT M DEGREE AN EXAM QDB4 26 Oct 2025). It carries 3 marks.

This question was also asked in

More questions from this paper

  1. Q1THIS IS QUESTION PAPER FOR THE SUBJECT "DEGREE LEVEL : DEEP LEARNING (COMPUTER BASED EXAM)" ARE YOU SURE YOU HAVE TO WR…
  2. Q2Figure question
  3. Q3Figure question
  4. Q4Which of the following statements best describes the primary representational advantage of a Multi-Layer Perceptron (ML…
  5. Q6Figure question
  6. Q7You are adapting a neural network that was originally designed for a 10-class image classification problem to perform a…
  7. Q8Figure question
  8. Q9Figure question
  9. Q10What is the value of the output of the hidden layer? (Answer correct upto two digits after the decimal)
  10. Q11What is the value of the cross entropy loss? (use natural log). (Answer correct upto two digits after the decimal)
  11. Q12Based on the above data, answer the given subquestions.
  12. Q13Use cross entropy loss and compute the gradient of W1 which is the weight between input and hidden layer. (consider upt…
  13. Q14Based on the above data, answer the given subquestions.
  14. Q15Based on the above data, answer the given subquestions.
  15. Q16Based on the above data, answer the given subquestions.
  16. Q17Based on the above data, answer the given subquestions.
  17. Q18After one iteration of gradient descent, will the new weight for W1 be the same as the previous weight?
  18. Q19After one iteration of gradient descent, will the new weight for W2 be the same as the previous weight?
  19. Q20Based on the above data, answer the given subquestions.
  20. Q21Based on the above data, answer the given subquestions.
  21. Q22Suppose you build a neural network with one hidden layer that uses sigmoid as an activation function and an output laye…