Deep Learning IIT Ropar Week 3 Nptel Answers
Are you looking for the Deep Learning IIT Ropar Week 3 NPTEL Assignment Answers? You’ve come to the right place! Access the most accurate and up-to-date solutions for your Week 3 assignment in the Deep Learning course offered by IIT Ropar.
Table of Contents
Deep Learning IIT Ropar Week 3 Nptel Assignment Answers (July-Dec 2025)
Question 1. What is the correct dimension of W2 in this setup?
a) 4×3
b) 3×4
c) 3×3
d) 3×1
Question 2. Which of the following loss–activation combinations are correctly used in AgroScan?
a) Softmax activation and squared error loss
b) Sigmoid activation and cross-entropy loss
c) Softmax activation and cross-entropy loss
d) Linear activation and cross-entropy loss
Question 3. Compute the pre-activation values for the hidden layer for the input x = [0.2, 0.6, 0.1, 0.5] with w1 = [[1 0 -1], [0 1 1], [1 0 1]]
a) [0.9, 0.7, 0.8]
b) [0.8, 0.9, 0.7]
c) [0.8, 0.7, 0.9]
d) [0.7, 0.9, 0.8]
Question 4. In the AgroScan model, after performing the forward pass, z = [0.96, −0.27, 1.19]. What class will the network predict after applying softmax?
a) Healthy
b) Pest-infected
c) Nutrient-deficient
d) All classes have equal probability
Question 5. In a classification task, the true label is “Pest-infected” and the predicted probability for that class is 0.55. Compute the cross-entropy loss (rounded to two decimals).
Fill in the blank: _________
Question 6. Given softmax output [0.35, 0.55, 0.10], what is the predicted class in the AgroScan model?
a) Healthy
b) Pest-infected
c) Nutrient-deficient
d) None of the above
Question 7. How many learnable parameters are there in the above neural network in total?
a) 12
b) 15
c) 21
d) 24
Question 8. If the hidden layer used tanh instead of sigmoid, what would change?
a) Activation outputs could be negative
b) Output layer would become invalid
c) Pre-activations would change
d) Number of parameters would increase
Question 9. Which of the following can lead to higher cross-entropy loss during training?
a) Predicting low probability for the true class
b) Predicting uniform probabilities for all classes
c) Predicting 1.0 for the correct class
d) Predicting a class different from the true one
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Question 10. Which of the following are always true for softmax output?
a) Output values lie in 0, 1
b) Output values sum to 1
c) Softmax is invariant to the order of inputs
d) Softmax is sensitive to relative magnitudes of logits
Question 11. Why is the output activation in LeafYield chosen to be linear?
a) Because softmax is unsuitable for regression
b) Because we need raw values, not probabilities
c) Because sigmoid would squash the output range
d) All of the above
Question 12. What is the final output if W3 = [1 1 −1] and h[2] = [0.73, 0.90, 0.91]?
Course Link: Click Here
a) 0.70
b) 0.71
c) 0.72
d) 0.73
Question 13. Which loss function is most appropriate for LeafYield?
a) Cross-entropy
b) Binary cross-entropy
c) Mean squared error
d) Kullback–Leibler divergence
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Question 14. Given the number of neurons and biases, compute the total number of learnable parameters in LeafYield.
Fill in the blank: __________
Question 15. Given linear output activation and final pre-activation value = 3.75, what will be the model’s prediction?
Fill in the blank: __________
Question 16. Which of the following statements are true about the activation functions used in LeafYield?
a) tanh outputs values between -1 and 1
b) sigmoid outputs values between 0 and 1
c) linear activation has no bounds
d) sigmoid is more centered around zero than tanh
View Answers
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Question 17. Given x = [0.1, 0.5, 0.3, 0.7, 0.2]ᵀ and W1 as given, what is a[1] (pre-activation vector)?
a) [0.6, 1.2, 0.5, 1.5]
b) [0.6, 1.2, 0.3, 1.5]
c) [0.6, 1.3, 0.5, 1.5]
d) [0.6, 1.3, 0.3, 1.5]
Question 18. Apply tanh activation to the above vector and round to 2 decimals.
a) [0.54, 0.83, 0.46, 0.91]
b) [0.54, 0.83, 0.29, 0.91]
c) [0.54, 0.86, 0.46, 0.91]
d) [0.54, 0.86, 0.29, 0.91]
Question 19. The LeafYield model was tested on 5 plant samples. What is the Mean Squared Error?
a) 0.12
b) 0.15
c) 0.25
d) 0.30
Question 20. Which layer’s weights will receive gradients first during backpropagation?
a) Output layer
b) Hidden layer 2
c) Hidden layer 1
d) Input layer
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Question 21. Which activation is best suited for the PlayPredict classification task?
a) Linear activation
b) Sigmoid activation
c) Tanh activation
d) Softmax activation
Question 22. What is the predicted class if the output vector contains [0.2, 0.1, 0.6, 0.1]?
a) Defender
b) Midfielder
c) Forward
d) Goalkeeper
Question 23. Given logits: [1.2, 0.8, 2.0, 1.0], compute softmax output (rounded to 2 decimals)
a) [0.22, 0.14, 0.48, 0.17]
b) [0.21, 0.14, 0.47, 0.17]
c) [0.28, 0.16, 0.58, 0.27]
d) [0.51, 0.54, 0.47, 0.17]
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Question 24. Why use tanh in the hidden layer?
a) Keeps outputs in [0, 1]
b) Introduces non-linearity
c) Allows negative outputs
d) Helps output softmax
Question 25. Given x = [0.5, 0.3, 0.2, 0.1, 0.4]ᵀ and W1 as given, what is the pre-activation a[1]?
a) [0.9, 0.9, 0.8, 0.7]
b) [0.9, 0.8, 1.1, 0.6]
c) [0.9, 0.9, 1.1, 0.7]
d) [1.0, 0.9, 1.1, 0.8]
Question 26. Compute the hidden vector h[1] from the above a[1] using tanh.
a) [0.72, 0.72, 0.66, 0.60]
b) [0.72, 0.66, 0.80, 0.54]
c) [0.72, 0.72, 0.80, 0.60]
d) [0.76, 0.72, 0.80, 0.66]
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Question 27. The PlayPredict model produced ŷ = [0.1, 0.7, 0.1, 0.1]. What is the categorical cross-entropy loss (rounded to 3 decimals)?
a) 0.105
b) 0.357
c) 0.500
d) 0.845
Question 28. When bias is used in both layers, what is the total number of learnable parameters in PlayPredict?
Fill in the blank: __________
Question 29. Which of the following can be tanh activation output?
a) [–0.9, 0.2, 0.8, –0.4]
b) [1.5, –1.2, 2.3, 0.5]
c) [0.0, 1.0, 2.0, –1.0]
d) [–2, –1, 0, 1]
Deep Learning IIT Ropar Week 3 Nptel Assignment Answers (Jan-Apr 2025)
1) How many parameters (including biases) are there in the entire network?
View Answer
2) Suppose all elements in the input vector are zero, and the corresponding true label is also 0. Further, suppose that all the parameters (weights and biases) are initialized to zero. What is the loss value if the cross-entropy loss function is used? Use the natural logarithm (ln).
View Answer
3) Choose the correct dimensions of W₁ and a₁
4) How many learnable parameters (including bias) are there in the network?
View Answer
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
5) We have a multi-classification problem that we decide to solve by training a feedforward neural network. What activation function should we use in the output layer to get the best results?
a) Logistic
b) Step function
c) Softmax
d) Linear
6) Which of the following statements about backpropagation is true?
a) It is used to compute the output of a neural network.
b) It is used to optimize the weights in a neural network.
c) It is used to initialize the weights in a neural network.
d) It is used to regularize the weights in a neural network.
7) Given two probability distributions p and q, under what conditions is the cross-entropy between them minimized?
a) All the values in p are lower than corresponding values in q
b) All the values in p are higher than corresponding values in q
c) p = 0 (0 is a vector)
d) p = q
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
8) Given that the probability of Event A occurring is 0.18 and the probability of Event B occurring is 0.92, which of the following statements is correct?
a) Event A has a low information content
b) Event A has a high information content
c) Event B has a low information content
d) Event B has a high information content
9) What is the predicted output for the given input x after doing the forward pass?
View Answer
10) Compute and enter the loss between the output generated by input x and the true output y.
View Answer
Deep Learning IIT Ropar Week 3 Nptel Assignment Answers (JULY – DEC 2024)
Q1. Choose the correct dimensions of W1
and a1
W1∈R6×3
W1∈R3×6
W1∈R1×6
a1∈R6×1
a1∈R1×6
a1∈R1×1
Answer: W1∈R6×3 , a1∈R6×1
Q2. How many learnable parameters(including bias) are there in the network?
Answer: 62
For answers or latest updates join our telegram channel: Click here to join
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Q3.Which of the following loss functions can be used for the classification problem?
Means Squared error
Cross entropy
Both Mean Squared Error and Cross Entropy loss can be used
Answer: Cross entropy
Q4. We have a multi-classification problem that we decide to solve by training a feedforward neural network. What activation function should we use in the output layer to get the best results?
Logistic
Step function
Softmax
linear
Answer: Softmax
For answers or latest updates join our telegram channel: Click here to join
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Q5.We have data x with the following labels y = [‘car’, ‘bus’, ‘bike’, ‘car’, ‘bus’, ‘car’, ‘bus’, ‘bike’, ‘car’]. Which of the following distribution will give the lowest cross-entropy loss with y? (Distribution is given in the following order [‘car’, ‘bus’,‘bike’])?
[0.49,0.30,0.21]
[0.44,0.33,0.22]
[0.50,0.30,0.20]
[0.44,0.39,0.19]
Answer: [0.44,0.39,0.19]
Q6. Which of the following statements about backpropagation is true?
It is used to compute the output of a neural network.
It is used to initialize the weights in a neural network.
It is used to optimize the weights in a neural network.
It is used to regularize the weights in a neural network.
Answer: It is used to optimize the weights in a neural network.
For answers or latest updates join our telegram channel: Click here to join
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Q7.Given two probability distributions p and q, under what conditions is the cross entropy between them minimized?
All the values in p are lower than corresponding values in q
All the values in p are lower than corresponding values in q
p=0 (0 is a vector)
p=q
Answer: p=q
Q8.Given that the probability of Event A occurring is 0.80 and the probability of Event B occurring is 0.20, which of the following statements is correct?
Event A has a low information content
Event A has a high information content
Event B has a low information content
Event B has a high information content
.
Answer: Event A has a low information content, Event B has a high information content
For answers or latest updates join our telegram channel: Click here to join
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Q9.What is the predicted output for the given input x after doing the forward pass? (Choose the option closest to your answer)
Answer: 4.955
Q10.Compute and enter the loss between the output generated by input x
and the true output y
Answer: 9.272
For answers or latest updates join our telegram channel: Click here to join
These are Deep Learning IIT Ropar Week 3 Nptel Assignment Answers
Check here all Deep Learning IIT Ropar Nptel Assignment Answers : Click here
For answers to additional Nptel courses, please refer to this link: NPTEL Assignment Answers