SRM VALLIAMMAI ENGINEERING COLLEGE

(An Autonomous Institution)

SRM Nagar, Kattankulathur – 603 203

DEPARTMENT OF
ELECTRICAL AND ELECTRONICS ENGINEERING

QUESTION BANK

II SEMESTER

1916210 - SOFT COMPUTING TECHNIQUES

Regulation – 2019

Academic Year 2019 – 20

Prepared by

Dr. G.Udhayakumar,Asso. Prof./EEE

 SRM VALLIAMMAI ENGINEERING COLLEGE
(An Autonomous Institution)
SRM Nagar, Kattankulathur – 603 203.

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

QUESTION BANK
SUBJECT : 1916210 - SOFT COMPUTING TECHNIQUES

SEM / YEAR: II/I M.E(PSE)

UNIT I - INTRODUCTION AND EVOLUTIONARY PROGRAMMING

Introduction to intelligent systems- Soft computing techniques- Conventional Computing versus
Swarm Computing - Classification of meta-heuristic techniques - Properties of Swarm intelligent
Systems - Application domain - Discrete and continuous problems - Single objective and multiobjective
problems. Evolutionary programming – Genetic algorithms & Genetic programming -
Particle Swarm Optimization- Ant colony Optimization
PART – A
BT Course
Q.No Questions Competence
Level Outcome
1. What is soft computing? BTL 1 Remember CO1
2. Compare soft computing vs. hard computing. BTL 2 Understand CO1
3. Classify the various types of soft computing techniques BTL 3 Apply CO1
4. Define single-objective optimization. BTL 1 Remember CO1
5. Define multi-objective optimization. BTL 1 Remember CO1
6. Define Intelligent System BTL 1 Remember CO1
7. Define Evolutionary Computation BTL 1 Remember CO1
8. What are the basic variations of Evolutionary Algorithm BTL 3 Apply CO1
9. Generalize the characteristics of intelligent systems BTL 2 Understand CO1
10. State the principles for an ideal multi-objective optimization BTL 2 Understand CO1
In what way if-then rules are used for multiobjective CO1
11. BTL 2 Understand
optimization?
State few advantages and disadvantages of Genetic CO1
12. BTL 2 Understand
Algorithm.
13. Mention the role of fitness function in Genetic Algorithm. BTL 2 Understand CO1
14. How is Genetic Algorithm differ from traditional algorithm? BTL 4 Analyse CO1
15. Write the applications of PSO? BTL 2 Understand CO1
16. Mention the advantages of Particle Swarm optimization BTL 2 Understand CO1
Compare and Contrast—Genetic Algorithm and Particle CO1
17. BTL 4 Analyse
Swarm Optimization
18. What are the characteristics of Ant Colony Optimization BTL 2 Understand CO1
19. State the various applications of ACO BTL 2 Understand CO1
Justify Swarm intelligence is superior to conventional CO1
20. BTL 4 Analyse
computing algorithm.
PART – B
1. Explain the classical methods for handling multi-objective (13) BTL 3 Apply CO1
 optimization problems.
Explain the procedure involved in using Evolutionary (13) CO1
2. BTL 1 Remember
programming for multi-objective problems.
(i) Explain the classification of Metaheuristic Techniques (6) BTL 2 Understand CO1
3.
(ii) Explain the Preparatory Steps of genetic programming. (7)
With a new flowchart, explain the Executional steps of (13) BTL 1 Remember CO1
4.
genetic programming.
Draw the functional block diagram of swarm intelligence (13) BTL2 Understand CO1
system, explain the role of sub blocks. Also represent the
5.
properties and application domain of swarm intelligence
system.
Implement a vehicle routing problem using the concept of (13) BTL 6 Create CO1
6.
particle swarm optimization.
Enumerate the procedure involved in using Genetic (13) BTL 2 Understand CO1
7.
Algorithm for optimizing controller parameters
With a neat flowchart, explain the algorithm of Genetic (13) BTL 1 Remember CO1
8.
Algorithm.
Implement Optimization of Traveling Salesman Problem (13) BTL 6 Create CO1
9.
using Genetic algorithm approach.
(i) What are the properties of Swarm intelligent Systems (5) BTL 2 Understand CO1
10.
(ii) Discuss the principles of multi-objective optimization (8)
With a neat flowchart, explain the algorithm of particle (13) BTL 1 Remember CO1
11.
swarm optimization.
With a neat flowchart, explain the algorithm of Ant Colony (13) BTL 1 Remember CO1
12.
Optimization
(i) Discuss the behavior of real ants and compare it with (7) BTL 2 Understand CO1
artificial ants.
13.
(ii) Differentiate the features of soft computing and hard
computing. (6)
Compare and Contrast—Genetic Algorithm, Particle Swarm (13) BTL 4 Analyse CO1
14.
Optimization and Ant Colony Optimization
PART – C
Implement a vehicle routing problem using the concept of (15) CO1
1. BTL 6 Create
particle swarm optimization.
Ant colony optimization algorithm is best suited for protein- (15) BTL 5 Analyse CO1
2.
folding problems—Justify
Assume a typical control problem of yours and create the (15) CO1
3. BTL 6 Create
various steps involved in finding a solution using GA
(15) BTL 3 Apply CO1
Let a function be defined on the interval
4. [0,31]. Apply Genetic Algorithm for determining the
maximum of the given function (Assume suitable missing
data)

UNIT II - FUZZY LOGIC SYSTEM

Introduction to crisp sets and fuzzy sets- basic fuzzy set operation and approximate reasoning.
Introduction to fuzzy logic modeling and control- Fuzzification inferencing and defuzzification-
Fuzzy knowledge and rule bases-Fuzzy modeling and control schemes for nonlinear systems.
Self-organizing fuzzy logic control- Fuzzy logic control for nonlinear time delay system. Case
study – Familiarization of FLC Tool Box.
PART – A
 BT Course
Q.No Questions Competence
Level Outcome
1. In What context fuzzy systems are used? BTL 1 Remember CO2
2. Mention the limitations of Fuzzy system? BTL 2 Understand CO2
3. Compare fuzzy set and crisp set? BTL 2 Understand CO2
4. Mention some applications of Fuzzy logic. BTL 3 Apply CO2
What is cardinality of a Fuzzy set? Whether a power set can BTL 1 Remember CO2
5.
be formed for a fuzzy set?
What is an empty Fuzzy set and What do you mean by BTL 1 Remember CO2
6.
height of a Fuzzy set?
7. Explain Centre of gravity method of defuzzification. BTL 1 Remember CO2
8. Explain Fuzzy compliment and Fuzzy relation ? BTL 1 Remember CO2
9. Classify the different Fuzzy relation operation? BTL 1 Remember CO2
10. Draw the block diagram of a Fuzzy logic system. BTL 1 Remember CO2
11. Define Fuzzy inference System? BTL 1 Remember CO2
12. Define a Fuzzy Cartesian product. BTL 1 Remember CO2
How is fuzzy relation converted into a crisp relation using BTL 1 CO2
13. Remember
lamda-cut process?
Explain the difference between conventional control and BTL 2 CO2
14. Understand
fuzzy control system.
15. Classify the methods of defuzzification? BTL 1 Remember CO2
16. What is the difference between fuzzy logic and binary logic? BTL 2 Understand CO2
17. Write the classifications of Fuzzy Logic control? BTL 1 Remember CO2
18. Illustrate any one defuzzification method with an example? BTL 3 Apply CO2
19. What is approximate reasoning? BTL 1 Remember CO2
20. Summarize Fuzzy rule and Fuzzy linqistic variable? BTL 4 Analyse CO2
PART – B
(i)Explain with neat block diagram the various components (8) CO2
BTL 1
1. of a fuzzy logic system (5) BTL 1
Remember
(ii)Describe shortly on Centroid method
Describe the self-organizing Fuzzy Logic Control scheme (13) CO2
2. with a suitable example. Mention its advantages over fuzzy BTL 3 Apply
logic controller.
Write short notes on implementation of Fuzzy Logic Control (13) CO2
3. BTL 6 Create
for nonlinear time delay system.
(i) Discuss the methods of aggregation of fuzzy rules. (7) CO2
4. (6)
BTL 1
Remember
(ii) Write short notes on fuzzy propositions.
(i) Differentiate between mamdani FIS and sugeno FIS. (5) CO2
(ii) The membership functions for the linguistic variables (8)
“tall” and “short” are given below.
BTL 2 Understand

5.
BTL 5 Evaluate

Describe the membership functions for the following

linguistic phrases. Very tall, fairly tall, not very short
Explain the features and benefits of the following Fuzzy CO2
6. BTL 4 Analyse
propositions (7)
 (i) Unconditional and qualified propositions (6)
(ii) Conditional and Unqualified propositions
Explain the different types of membership function used in (13) CO2
7. BTL 3 Apply
fuzzification process?
Illustrate with suitable examples explain canonical rule (13) CO2
8. BTL 3 Apply
formulation of fuzzy rule base
The results of three implication processes arc as shown in (13) CO2
fig. Find the aggregated output and the defuzzified output
using the (1) Center of gravity (2) Center of sums and (3)
Weighted average methods

9. BTL 5 Evalaute

Compare the following two fuzzy relations R1and R2 Using (13) CO2
max-min and max average compositions:

R1 y1 y2 y3 y4
x1 0.3 0 0.7 0.3
x2 0 1 0.2 0
R2 z1 z2 z3
10. BTL 5 Evalaute
y1 0.1 0.2 0.4
y2 0.8 0.3 1

y3 0.7 0.9 0.6

y4 1 0.2 0.1

Develop a typical case study in which Fuzzy logic is used as (13) CO2
11. BTL 6 Create
a model and controller
The discretized membership functions for a transistor and a (13) CO2
resistor are given below:
uT={ 0/0 + 0.2/1 + 0.7/2 + 0.8/3 + 0.9/4 + 1/5}
uR={ 0/0 + 0.1/1 + 0.3/2 + 0.2/3 + 0.4/4 +0.5/5}
12. BTL 5 Evalaute
Find the following (a) Algebraic sum (b) Algebraic product
(c) Bounded sum (d) Bounded difference

(i) Explain self organizing FLC. Mention its advantages (7) CO2
BTL 2
over fuzzy logic controller (6)
13. BTL 2 Understand
(ii) Sketch the block diagram of Fuzzy Logic Controller for
a nonlinear process.
14. (i) Find the algebraic product of two fuzzy sets ‘A’ and ‘B’ (7) BTL 3 Apply CO2
 where A={(3, 0.8)(5, 0.6)(6, 0.9)} and B={(3, 0.7)(4,
0.8)(5, 0.3)} (6)
(ii) Find the algebraic sum of two fuzzy sets A={(3, 0.5) (5,
1)(7, 0.6)} and ‘B’ and B={(3, 1)(5, 0.6)}

PART – C
Given a conditional and qualified Fuzzy proposition ‘P’ of (15) CO2
the form.
P: If x is A, then y is B is S where ‘ S’ is fuzzy truth qualifier
1. BTL 6 Create
and a fact is in the form” x is A” .We want to make an
inference in the form “y is B “ Design a method based on the
truth- value restrictions for getting the inference
Design the general scheme for a Fuzzy controller. How (15) Create CO2
different modules are interconnected? Deploying the above
2. BTL 6
how will you solve the problem of stabilizing the inverted
pendulum
Consider four travel packages offered by Celtic, Club (15) Evaluate CO2
Mahindra, Metro and Himalaya travels. We want to choose
one. Their costs are INR 100,000, INR 200,000 , INR
150,000 and INR 175,000. Their travel time in hours are
150, 200, 100 and 125 respectively. They are viewed as
3. BTL 5
interesting with degress 0.4, 0.3, 0.6 and 0.5. Define your
own fuzzy set of acceptable travel times. Then determine the
fuzzy sets of interesting travel packages whose cost and
travel times are acceptable and use this set to choose one of
your packages.
Design a fuzzy logic controller for a non-linear system of (15) CO2
4. BTL 6 Create
your choice with a case study

UNIT III - ARTIFICIAL NEURAL NETWORKS

Neuron- Nerve structure and synapse- Artificial Neuron and its model- activation functions-
Neural network architecture- single layer and multilayer feed forward networks- Mc Culloch Pitts
neuron model- perceptron model- Adaline and Madaline- multilayer perception model- back
propogation learning methods. Counter propagation network- architecture- functioning &
characteristics of counter Propagation network- Hopfield/ Recurrent network configuration-
Adaptive Resonance Theory. Case study – Familiarization of NN Tool Box.
PART – A
BT Course
Q.No Questions Competence
Level Outcome
1. Distinguish between artificial neuron & biological neuron. BTL 2 Understand CO3
2. Explain the factors affecting back propagation training? BTL 4 Analyse CO3
3. Name some activation functions used in ANN? BTL 1 Remember CO3
4. Explain a single layer net and multilayer net? BTL 1 Remember CO3
5. What is a back propagation NN? BTL 1 Remember CO3
Draw the basic model of Adaline network and Madaline CO3
6. BTL 1 Remember
Network
7. Sketch the model of artificial neuron. BTL 3 Apply CO3
8. Evaluate the learning rate and its function BTL 1 Remember CO3
For derivative -based learning procedure why a CO3
9. BTL 3 Apply
sigmoidal function is used instead of a step
 function?
10. Define an artificial neural network. BTL 1 Remember CO3
11. Draw a network for solving Exclusive OR problem. BTL 1 Remember CO3
12. State the function of synapse. BTL 1 Remember CO3
Justify why Artificial Neural Network is called adaptive BTL 6 CO3
13. Create
system during training
14. What is meant by winner take all? BTL 1 Remember CO3
15. What is the purpose of Hopfield Network? Give an example? BTL 2 Understand CO3
16. Why Hopfield network is called as recurrent neural network? BTL 4 Analyse CO3
17. Conclude the limitations of competitive learning? BTL 4 Analyse CO3
18. Give the characteristics of counter propagation network BTL 2 Understand CO3
19. Explain stability-plasticity dilemma? BTL 1 Remember CO3
20. What are the properties of adaptive resonance theory? BTL 2 Understand CO3
PART – B
1. Write short notes on Adaline and Madaline networks (13) BTL1 Remember CO3
Explain with a neat diagram the neural network (13) BTL 1 Remember CO3
2.
architecture of multilayer feed forward network.
3. Draw and describe the structure of a biological neuron. (13) BTL 1 Remember CO3
Using McCulloh-Pitts neuron model, design a neural network (13) BTL6 Create CO3
4.
for 2-input OR functions.
Explain the working of back propagation neural (13) BTL 4 Analyse CO3
5.
network with neat architecture and flowchart.
Using Adaline network to train ANDNOT function with (13) BTL6 Create CO3
6.
bipolar inputs and targets. Perform 2 epochs of training.
Determine the weights of a single layer perceptron for (13) BTL 5 Evaluate CO3
7. implementing the AND function. Consider the inputs and
targets to be bipolar and α=1
Explain the single perceptron with its learning algorithm and (13) BTL 1 Remember CO3
8.
its seperability and convergence property
Develop and describe with a neat diagram the counter (13) BTL 6 Create CO3
9.
propagation network learning algorithm.
Consider the following full counter propagation network (13) BTL 5 Evaluate CO3
(CPN) shown in below fig. (a) using input pair x(1,1)y=(0,1);
perform first phase of training (One step only).Find the
activation of the cluster layer units and update the weights
using learning rates of 0.3.

10.

Explain basic structure, working and analysis of adaptive (13) BTL 4 Analyse CO3
11.
resonance theory(ART-1)
Draw the architecture of full counter propagation network (13) BTL 3 Apply CO3
12. and represent the active units in the first and second phase of
counter propagation training. Explain how training a counter
 propagation network occurs in two phases.
Draw the architecture of Hopfield net.design Hopfield net (13) BTL 6 Create CO3
for 4 bit bipolar pattern The training pattern are I sample
13. S1[1,1,-1,-1] II sample S2[-1,1,-1,1] III sample S3[-1,-1,-1,-
1,1]

Describe with a neat diagram the architecture of recurrent (13) BTL 3 Apply CO3
14.
network to perform XOR task with two inputs.
PART – C
Can a two input Adaline compute the XOR function? Analyse the (15) BTL 4 Analyse CO3
1. XOR function using Madaline?
Find new weight when the net illustrated in given figure is (15) BTL 5 Evaluate CO3
presented the input pattern (0, 1) and target output is 1. Use a
learning rate α = 0.25 and the binary sigmoid activation
function.

1
1
0.4
-0.3
2. x1 0.7
z1 0.5
-0.2
y
-0.4
x2 z2 0.1
0.3
0.6

Consider the following full counter propagation network (15) BTL 5 Evaluate CO3
(CPN) shown in below fig. (a) using input pair x(1,1) &
y=(0,1); perform phase I & II of training update the weights
using learning rates of 0.3. (One step only)

3.

Consider an ART 1 network with four F1 units and three F2 (15) BTL 5 Evaluate CO3
units. Assume the initial weights as follows:
Bottom up weights (bij) :
{{0.67, 0, 0.3}, {0, 0, 0.3}, {0, 0, 0.4}, {0, 0.44, 0.5}} and
4.
Top down weights (tij) :
{{1, 0, 0, 1}, {0, 1, 0, 1}, {1, 1, 1, 0}}.
Determine the new weight matrices after the vector (0,
1, 0, 1) is presented, if the vigilance parameter is given as
 0.25 and 0.7
UNIT IV - HYBRID CONTROL SCHEMES
Fuzzification and rule base using ANN – Neuro fuzzy systems-ANFIS – Fuzzy Neuron -
Optimization of membership function and rule base using Genetic Algorithm. Familiarization of
ANFIS Tool Box.
PART – A
BT Course
Q.No Questions Competence
Level Outcome
1. List few applications of hybrid fuzzy GA systems. BTL 2 Understand CO4
2. List few applications of Neuro fuzzy systems. BTL 2 Understand CO4
What are the parameters selected when implementing CO4
3. BTL 3 Apply
Fuzzy Logic Control using MATLAB
What are the transfer functions available in MATLAB neural CO4
4. BTL 3 Apply
network toolbox.
Classify the shapes of the membership function available in CO4
5. BTL 4 Analyse
fuzzy logic tool box .
Name any three commercial software used for soft CO4
6. BTL 4 Analyse
computing techniques.
Give the defuzzification methods available in MATLAB CO4
7. tool box And What are the inference methods available BTL 4 Analyse
in MATLAB tool box?
8. What is the purpose of toolboxes in matlab? BTL 2 Understand CO4
9. Write a few NN readily available in Mat lab tool box. BTL 2 Understand CO4
10. Compare Fuzzy Processing and Neural Processing BTL 4 Analyse CO4
11. What are the classifications of Neuro Fuzzy Hybrid Systems BTL 1 Remember CO4
State the limitations of neural networks and fuzzy systems CO4
12. BTL 2 Understand
when operated individually
13. List the various types of hybrid systems. BTL 2 Understand CO4
14. Mention the properties of neuro-fuzzy hybrid systems BTL 2 Understand CO4
15. Mention the characteristics of neuro-fuzzy hybrid systems BTL 4 Analyse CO4
16. What are advantages of Genetic Fuzzy Hybrids system BTL 2 Understand CO4
What are the objectives of optimization in fuzzy rule based CO4
17. BTL 4 Analyse
system
Name the approaches used for Genetic Learning of Rule CO4
18. BTL 2 Understand
Bases
19. List the Advantages of Neuro·Genetic Hybrids BTL 2 Understand CO4
Distinguish between tuning and learning problems of fuzzy CO4
20. BTL 4 Analyse
system.
PART – B
What is called ANFIS? Draw the architecture of ANFIS (13) CO4
1. BTL 1 Remember
network and represent the role of different layers.
With suitable block diagram, explain the principle involved (13) CO4
2. BTL 4 Analyse
in a liquid level controller using neurofuzzy technique.
Discuss with relevant diagrams and mathematical (13) CO4
expressions how a nonlinear system can be identified and
3. BTL 2 Understand
controlled using MATLAB Neural Network Tool box.
Choose appropriate example
Using MATLAB FIS tool box discuss how will you identify (13) CO4
4. BTL 3 Apply
and control the linear and nonlinear dynamic system
Using MATLAB ANFIS tool box discuss how will you (13) CO4
5. BTL 3 Apply
identify and control the linear and nonlinear dynamic system
 Using MATLAB Neural Network tool box discuss how will (13) CO4
6. you identify and control the linear and nonlinear dynamic BTL 3 Apply
system
Show how neuro fuzzy logic control can be used for washing (13) CO4
7. BTL 6 Create
machine applications
Design a generic algorithm to optimize the weights of a (13) CO4
8. neural network model while training an OR gate with 2 BTL 6 Create
bipolar inputs and 1 bipolar targets.
Write a MATLAB program train NAND gate with binary (13) CO4
9. inputs and targets(two input-one Output) using adaptive BTL 3 Apply
neuro-fuzzy hybrid technique.
How are generic algorithms utilized for optimizing the (13) CO4
10. BTL 5 Evaluate
weights in neural network architecture?
Explain in detail the concepts of fuzzy generic hybrid (13) CO4
11. BTL 1 Remember
systems
Give details on the various applications of neuro fuzzy (13) CO4
12. BTL 2 Understand
hybrid systems.
13. Explain the classifications of neuro-fuzzy hybrid systems? (13) BTL 1 Remember CO4
With suitable block diagram, explain the principle involved (13) CO4
14. BTL 6 Create
in a liquid level controller using neurofuzzy technique.
PART – C
Create a neuro and fuzzy controller for its application in (15) CO4
1. BTL 6 Create
inverted pendulum system.
Show how fuzzy logic control and genetic algorithm based (15) CO4
2. structural optimization can be used for plant control BTL 6 Create
applications
Assume a typical control problem of yours and explain the (15) CO4
3. various steps involved in finding a solution using Fuzzy and BTL 6 Create
GA.
4. Create a GA-Fuzzy Systems for Control of Flexible Robots (15) BTL 6 Create CO4
UNIT V - INTRODUCTION TO MACHINE LEARNING AND DEEP LEARNING
Linear Regression, Logistic Regression, Naive Bayes Classifier, kNN algorithm, Support Vector
Machines (SVMs) and Decision Trees, Random Forest Classifier, Deep Learning-Recurrent
Neural Networks and Convolutional Neural Networks and Reinforcement Learning: Markov
Decision processes (MDPs) and Q-learning
PART – A
BT Course
Q.No Questions Level
Competence
Outcome
1. Define Logistic regression BTL 1 Remember CO5
2. Why Is Logistic Regression Popular? BTL 2 Understand CO5
Which of the following distance measure do we use in case CO5
3. BTL 3 Apply
of categorical variables in k-NN?
4. How do we decide the value of "K" in KNN algorithm? BTL 4 Analyse CO5
5. Why is KNN algorithm called Lazy Learner? BTL 2 Understand CO5
What are the advantages and disadvantages of KNN CO5
6. BTL 2 Understand
algorithm?
7. What are the limitations of Random Forest Classifier? BTL 2 Understand CO5
8. Define SVM. BTL 1 Remember CO5
What are the problems with the following characteristics is CO5
9. BTL 4 Analyse
generally suited in decision tree learning?
10. Why is “Naive” Bayes naive? BTL 2 Understand CO5
 11. What are the properties of SVM. BTL 1 Remember CO5
12. Mention the advantages and disadvantages of SVM. BTL 2 Understand CO5
What is deep learning, and how does it contrast with other CO5
13. BTL 5 Evaluate
machine learning algorithms?
14. Differentiate between Perceptron and SVM BTL 4 Analyse CO5
15. How is a decision tree pruned? BTL 2 Understand CO5
16. What are the applications of SVM. BTL 2 Understand CO5
17. What are the Different Layers on CNN? BTL 1 Remember CO5
18. What are the Challenges of Reinforcement Learning BTL 2 Understand CO5
19. Name the Kernels used in SVM classification process. BTL 3 Apply CO5
20. Justify the use of Reinforcement Learning? BTL 5 Evaluate CO5
PART – B
1. Explain the Naïve Bayesian Classifier (13) BTL 1 Remember CO5
2. Discuss the Linear regression for machine learning (13) BTL 2 Understand CO5
With a neat flowchart, explain the working of k-NN (13) CO5
3. BTL 1 Remember
algorithm.
(i) What are the important objectives of machine learning? (3) CO5
4. BTL 4 Analyse
(ii) Discuss different examples of machine learnig. (10)
5. Discuss the Logistic regression for machine learning (13) BTL 2 Understand CO5
With neat flowgraph and explain the Pooling of Convolution (13) CO5
6. BTL 1 Remember
netork
7. Explain in detail the operation of Support Vector machines. (13) BTL 1 Remember CO5
Discuss the learning tasks and Q learning in the context of (13) CO5
8. BTL 4 Analyse
reinforcement learning
9. With neat algorithm and explain Random Forest Classifier (13) BTL 1 Remember CO5
Differentiate between linear SVM and Kernel SVM (5) BTL 2 Understand CO5
10.
Explain with different kernels used in SVM. (8) BTL 1 Remember
Detailed workflow for training and evaluating a deep (13) CO5
11. BTL 5 Evaluate
learning model
(i) Compare Reinforcement Learning vs. Supervised (5) CO5
Learning BTL 4 Analyse
12.
(ii) Applications of Reinforcement Learning (4) BTL 2 Understand
(iii)Challenges of Reinforcement Learning (4)
13. With a neat flowchart, explain the algorithm of SVM. (13) BTL 1 Remember CO5
Explain the components of Markov Decision Process and (13) CO5
14. BTL 5 Evaluate
how to find optimal policy.
PART – C
(i) Discuss the use of decision tree for classification problem (5) CO5
1. (ii) Describe the ID3 algorithm for decision tree learning (10) BTL 2 Understand
with example
Explain the architecture of Convolutional Neural Network (15) CO5
2. BTL 2 Understand
with an example
(i) Explain Hypothesis space search in decision tree (5) CO5
(ii) Explain inductive in decision tree learning (5)
3. BTL 1 Remember
(iii)Write a short note with diagram on Decision trees, which (5)
are non linear, non metric classifiers
For choice of your application, design and train the SVM (15) CO5
4. BTL 6 Create
network with different kernels and classify them.