IVB.
TECH– ISEMMID-II EXAMINATIONS
Course: DEEP LEARNING TECHNIQUES
Duration: 90 min
Date: Program : CSE
*************************************************
AnswerAllThreequestions Marks:30
Blooms
S.
Question Marks CO No. Taxonomy
No
Level
UNIT-1
a) Write down brief history and evolution of AI.
1 10 CO1 Remembering
b) Write down present and future scope of AI.
a) How random forests are related to Decision trees.
2 10 CO1 Understand
b) How is it possible to perform un-supervised
learning with Random Forest?
a) What are kernel methods in Deep learning?
3 Explain. 10 CO1 Apply
b) Explain the terms over fitting and Under fitting
In ML.
4 What is a Decision tree algorithm? Explain. 10 CO1 Apply
Discuss about Probabilistic modeling in detail(Naïve
5 10 CO1 Remember
Bayes Algorithm)
a) Explain how random forests give output for
6 classification and regression problems. 10 CO1 Analyze
b) Write the differences between RF and DT.
UNIT-2
Explain the difference between AI, ML and DL
1 10 CO2 Remembering
Explain the terms forward and backward
2 10 CO2 Understand
propagation in ML with example
a) Explain about biological vision and machine
3 vision. 10 CO2 Understand
b) How to improve Deep learning using weight
initialization.
a) Compare traditional machine learning
approaches with current deep learning
4 10 CO2 Apply
approaches.
b) Explain the deep learning network architecture.
a) Elaborate on various cost functions used in
5 10 CO2 Remember
training deep networks.
b) How to increase accuracy in deep networks.
a) Illustrate on computation representation of
6 10 CO2 Analyze
language in Human and Machine language.
b) Explain the forward propagation in Deep NN with
suitable example.
UNIT-3
a) Explain about the architecture of Keras.
1 10 CO3 Remember
b) Discuss about keras workflow.
a) Explain the anatomy of a neural network.
2 b) Explain the terms loss function and optimizers 10 CO3 Analyze
with respect to DL.
a) With a neat sketch, enumerate the concept of
3 10 CO3 Apply
the deep-learning software hardware stack.
b) Explain different types of neural networks.
