Roll No.

3533
B. Tech. 7th Semester (CSE)
Examination May, 2023
NEURAL NETWORKS
Paper: PCC-CSE-401-G

Time: Three hours ] [Maxinmum Marks : 75

Before answering the questions, candidates should ensure that they

have been supplied the correct and complete question paper. No
complaint in this regard, willbe entertained after examination.
Note: Attempt five questions in all, selecting one question
from each Section. Question No. 1 is compulsory.
All questions carries equal marks.
1. Explain the following : 2.5 x 6 = 15

(a) Application of Neural Networks

(b) Activation Function
(c) Describe Learning Factors

(d) Structure of Biological Neuron

(e) Auto Associative Memory
() Hetro Associative Memory
3533-)oÝ(P-3)(Q-9)(23) P.T.O.
 SECTION -A

2. Explain architecture of artificial Neural Network

Explain its three basic types of Neuron connection
architecture. 15

3. What is associative memory ? Explain its various types

in detail with suitable example. 15

SECTION-B

4. What do you mean by McCulloch Pitts Model ?

Explain this model by using AND function. 15

5. Explain the following: 15

(a) Gradient Descent Algorithm

(b) Linear Separability
SECTION-C

6. Explain Single Layer architecture of Perceptron

Network and implement AND Function using
Perceptron Network. 15

7. Describe the following : 15

(a) Supervised Learning

(b) Unsupervised Learning
(c) Reinforcement Learning
3533 -(P-3\Q-9)(23) (2)
 SECTION -D

8. Explain Hebb Model in detail 15

9 What is Competitive Learning ? How it is different

from conventional Learning ? Explain its significance
and usefulness in Neural Network 15

3533- -(P-3)(Q-9)(23) (3)

Roll No.

3533
B. Tech. 7th Semester (CSE)
Examination - May, 2023
NEURAL NETWORKS
Paper :PCC-CSE-401-G

[Maximum Marks:75
Time :Three hours ]
ansuering the questions, candidates should ensure that they
Before
question paper. No
have been supplied the correct and complete
examination.
complaint in this regard, will be entertained after
selecting one question
Note : Attempt five questions in all,
from each Section. Question No. 1 is compulsory.
All questions carries equal marks.
2.5 x 6= 15
1. Explain the following:
(a) Application of Neural Networks
(b) Activation Function

(c) Describe Learning Factors

(d) Structure of Biological Neuron
(e) Auto Associative Memory
() HetroAssociative Memory
P. T.O.
3533-)oP-3)(Q-9)23)
 SECTION-A

2. Explain architecture of artificial Neural Network

Explain its three basic types of Neuron connection
architecture. 15

3. What is associative memory ? Explain its various types

in detail with suitable example. 15

SECTION - B

4. What do you mean by McCulloch Pitts Model ?

Explain this model by using AND function. 15

5. Explain the following : 15

(a) Gradient Descent Algorithm

(b) Linear Separability

SECTION -C

6. Explain Single Layer architecture of Perceptron

Network and implemnent AND Function using
Perceptron Network. 15

7. Describe the following : 15

(a) Supervised Learning

(b) Unsupervised Learning
(c) Reinforcement Learning
3533 -(P-3)(Q-9)(23) (2)
 3533
B.Tech. (CSE) 7th Semester (G-Scheme)
Examination, December-2022
NEURALNETWORKS
Paper -PCC-CSE-401-G
[Maximum marks : 75
Time allowed: 3 hours]
is
Note: All questions carry equal marks. Ouestion no. I
compulsory. In addition to the compulsory question,
student will have to attempt four more questions se
lecting one questionfrom each unit.
1. Compulsory Question :
network.
(a) Discuss evolution of neural
Learning.
(b) Explainthe Comnpetitive
Obtain the output of neuron Y neuron having three
(c)
and weight are
input x, =1, x, = 2, and x, =3 Threshold and
w,= 1, w, = 1, w, = 2 by using
Sigmodial activation functions. -

capacity in
(d) Discuss the concept of Storage
Associative Memory.
Unit - I
Biological Neuron. Also
2. Explain the component of aequivalencies to artificial
focus on Biological neuron
neuron model.
Why it is used? Give
3. What is Activation Function?
in detail.
different types of activation function

2533-p. -9(22) [P.T.O.

 (2) 3533
Unit - II

function for
4. What is perceptron? Also realize if for OR
bipolar data.
5. Explain Linear Separability Concept by taking a suitable
using
example, also classify the output of OR function
it.
Unit- III
with
6. Derive Gradient Decent algorithm and compare it
generalized delta learning rule.
7. What is Learning? Explain its different types.
Unit- IV
Associative
8. Store the vector (1 1- 1 -1) in Auto
Network,And
(a) Find the Weight Matrix
(b) Test the net with input vector
(c) Test with one mistake in input
(d) Test with one missing in input
(e) Test with two missing in input
(f) Test with two mistake in input
9. What is Associative Memory? Explain Auto Associative
Memory with its architecture, training (insertion) and
testing (Retreival) Algorithm.
3533
Roll No.

3533
B. Tech. 7th Semester (CSE)
Examination February, 2022
NEURAL NETWORKS

Paper: PCC-CSE-401-G
Time : Three Hours ] [Maxinum Marks : 75
Before answering the questions, candidates should ensure that they
have been supplied. th» correct and complete question paper. No
complaint in this regard, will be entertained after exanination.
Note: Attempt five questions in all, selecting one question
from each Section. Question No. 1 is compulsory.
All questionscarry equal marks.
1. Explain the following : 5 x3 15
.

(a) Hebbian léarning.

(b) Reinforcement learning.
(c) Feedforward vs feedback Networks.
(d) What is the need of Activation Functions in
ANN ?

(e) Explain term Linear Separability classification.

3533-1900-(P-3)(Q-9)(22) P.T. O.
 SECTION- A

2. What are biological neurons ? How they resemble

artificial neuron models. Compare and contrast
biological neurons with Artificial Neural Networks. 15
3. (a) Explain various architecture models of ANN with
their corresponding advantages and

disadvantages. 8

(b) What are Activation functions and why we need

these functions in ANN ? Also write significance
of any three non-linear functions used in ANN. 7
SECTION- B

4. Discuss architecture of McCulloch Pitts Neural

Network model in detail. Also explain McCulloch Pitts
model to design logicnetworks of AND and OR.logic
function. 15

5. Explain perceptron network training architecture in

detail. Also, write implementation of AND function
using perceptron learning. 15

SECTION -C

6. (a) Differentiate between supervised and

1supervised learning.
(b) Explain Gradient Descent Algorithm in detail. 7

3533-1900-(P-3)(Q-9)(22) (2)
7. (a) Explain Error Back propagation algorithm in
detail.

(b) Write short note on Delta Learning rule.

SECTION D

8. (a) Explain different typeof Associative mnemories in

detail with example. 7

(b) Write storage and Retrieval algorithm for

associative memnory. 8

Explain bidirectional associative mnemory architecture,

its association encoding and decoding in detail. 15

3533-1900-(P-3)(Q-9)(22) (3)