ANN - PESIT South Campus

P.E.S. Istitute of Technology( Bangalore South Campus)
Hosur Road, ( 1Km Before Electronic City), Bangalore 560100.
Department of Electronics and Communication
SCHEME AND SOLUTION
_____I__ INTERNAL TEST
Faculty:
Subject:
Q.No.
1
2
Shwetha S Bhat
Semester: 7th Sem A,B,C
Artificial Neural Networks
Sub. Code:
What are Artificial Neural Networks?  page 1
What is Neuron Learning?  page 22
Write a brief on different types of learning.  page 22-25 , 1.4.1 to 1.4.3 in brief
Draw and name the terms used in general Neuron model and Biological model. 
fig 1.3 and 1.5 with its terms or use table 1.4
Explain the different types of Neural Net Architectures? 
1.3.1 to 1.3.5 with brief and diagrams for each
Marks
2
2
4
2
5x2
3
Give a brief on the following applications of ANN:
a. Forecasting  1.5.6
b. Function Approximation 1.5.5
c. Pattern Association 1.5.4
3+3+4
4
Define the following with examples:
a. Perceptron  2.1
b. Linear Separability  2.2 page 51
c. Linearly Non-Separable  2.2 page 52
3x2
5
6
Execute perceptron training algorithm on OR gate using initial weights w0 = 1, w1 = 1,
w2 = -1, learning rate: η = 1, with signum transfer function for classification.
4
What is the approximate choice of Learning rate and termination criteria in the Perceptron
Training Algorithm?  2.3.2 and 2.3.1
3+3
A fully connected feed-forward network has 5 source nodes, 2 hidden layers, the first
one with 4 neurons and the other with 2 neurons, and a single output neuron.
Construct an architectural graph of this network
4
What is Adaline linear element? Write the Adaline Training Algorithm and brief the steps
to solve a problem using the same.  2.5.2 and fig 2.9 with brief on the steps written in
class
Implement the perceptron training rule of the network using signum function with initial
weights W = [1, -1, 0, 0.5], learning rate: η = 1, for the 3 training pairs:
X1=[1,-2, 0, 1] , d1 = -1
X2 = [0, 1.5, -0.5, -1], d2 = -1
X3 = [-1, 1, 0.5, -1], d3 = 1
5
5