DR.
BABASAHEB AMBEDKAR TECHNOLOGICAL UNIVERSITY, LONERE
Regular Winter Examination – 2024
Course: BTech Branch : Electronics and Computer Engineering/ Electronics and Computer Science
Semester :VII
Subject Code & Name: BTECPC702 DEEP LEARNING
25333457
25333457
25333457
Max Marks: 60 Date:07/02/2025 Duration: 3 Hr.
Instructions to the Students:
1. Each question carries 12 marks.
2. Question No. 1 will be compulsory and include objective-type questions.
3. Candidates are required to attempt any four questions from Question No. 2 to Question No. 6.
4. The level of question/expected answer as per OBE or the Course Outcome (CO) on which the
question is based is mentioned in ( ) in front of the question.
5. Use of non-programmable scientific calculators is allowed.
6. Assume suitable data wherever necessary and mention it clearly.
(Level/CO) Marks
Q. 1 Objective type questions. (Compulsory Question) 12
1 What is the main component of a simple artificial neural network? (L1/CO1) 1
a) Neurons b) Layers c) Weights d) All of the above
2 (L1/CO2) 1
25333457
25333457
25333457
Which of the following is NOT an activation function used in CNNs?
a) Sigmoid b) ReLU c) Tanh d) Convolution
3 High bias indicates (L1/CO3) 1
a) Overfitting b) Underfitting c) High variance d) Low variance
4 The gates in LSTM include: 1
a) Update and reset gates b) Input, forget, and output gates
(L1/CO4)
c) Weight and bias gates d) Memory gates only
5 Image denoising is a task where the model:
a) Adds noise to images b) Removes noise from images
(L1/CO5) 1
c) Detects objects in images d) Segments objects in images
6 What function do activation functions typically perform in a neural
network?
(L1/CO1) 1
a) Scale the inputs b) Add non-linearity
c) Update weights d) Optimize gradients
25333457
25333457
25333457
7 What is the purpose of a filter (or kernel) in a convolutional layer?
a) Extract features from the input b) Pool data
(L1/CO2) 1
c) Perform activation d) Reduce dimensionality
8 Which domain commonly uses transfer learning?
a) Image recognition b) Natural language processing
(L1/CO3) 1
c) Speech recognition d) All of the above
9 Which of the following is a sequential model?
a) CNN b) RNN c) Decision Tree d) SVM (L1/CO4) 1
10 A GAN consists of two main components:
a) Encoder and decoder b) Generator and discriminator
(L1/CO5) 1
c) Classifier and segmenter d) Autoencoder and transformer
Page 1
11 Which of the following is NOT an optimization algorithm? (L1/CO1) 1
a) Adam b) RMSProp c) Gradient Descent d) ReLU
12 Which activation function is most commonly used in CNNs? (L2/CO2) 1
a) Sigmoid b) Tanh c) ReLU d) Softmax
Q. 2 Solve the following. 12
25333457
25333457
25333457
A) Illustrate the working principle of Simple Artificial Neural Network (L2/CO1) 6
With suitable diagram.
B) Discuss Forward Propagation and Backward Propagation Learning (L2/CO1) 6
in Deep Neural Network.
Q.3 Solve the following. 12
A) Explain building blocks of CNN with suitable labelled diagram. (L2/CO2) 6
B) Discuss Different CNN Architectures in detail. (L2/CO2) 6
Q. 4 Solve Any Two of the following. 12
A) Define the Transfer Learning and identify the need of Transfer (L1,L2/CO3) 6
25333457
25333457
25333457
Learning in Deep Neural Network.
B) Differentiate Between Early Stopping and Drop Out in Deep Neural (L1/CO3) 6
Network.
C) Illustrate Data Augmentation in Deep Neural Network. (L2/CO3) 6
Q.5 Solve Any Two of the following. 12
A) Illustrate the working of Recurrent Neural Network with neat (L2/CO4) 6
labelled diagram.
B) Summarize the working principle of LSTM with neat labelled (L2/CO4) 6
diagram.
C) Outline the working concept of GRU-Gated Recurrent Unit with (L2/CO4) 6
suitable diagram.
25333457
25333457
25333457
Q. 6 Solve Any Two of the following. 12
A) Elaborate Neural Style Transfer Network in detail. (L2/CO5) 6
B) Discuss Variational Auto Encoder. (L1/CO5) 6
C) Write Short note on Generative Adversarial Networks (L1/CO5) 6
*** End ***
Page 2
