Foundations of AI Course Code: CE-638

Assignment-2 (03/11/2022)
Basics of Artificial Neural
Networks
by
Dr.Upasna Singh

Department of Computer Science and Engineering

Defence Institute of Advanced Technology, Girinagar
Pune-411025
Instructions to be followed:
• Module 1: Hard copy needs to be submitted on or before 15/11/2022 by
4.15 pm
• Module 2: Hard Copy needs to be submitted on or before 20/11/2022 by
4.15 pm

MODULE 1
Q1. Solve the three simple classification problems shown in figure given below
by drawing a decision boundary. Find weight and bias values that result in
single-neuron perceptrons with the chosen decision boundaries. Initial weights
can be considered as (a) [-2 1] (b)[0 -2] (c) [2 -2]

Q2. Convert the classification problem defined below into an equivalent

problem definition consisting of inequalities constraining weight and bias
values.
X{[0 2], [1 0], [0 -2], [2 0]}; Y{1, 1, 0, 0}

Q3. We have a classification problem with four classes of input vector. The four
classes areClass 1: {[1 1], [1 2]}; Class 2: {[2 -1], [2 0]}; Class 3: {[-1 2], [-2
1]}; Class 4: {[-1 -1], [-2 -2]}.Design a perceptron network to solve this
problem.

Q4. Solve the following classification problem with the perceptron rule. Apply
each input vector in order, for as many repetitions as it takes to ensure that the
problem is solved. Draw a graph of the problem only after you have found a
solution. Consider initial weight and bias as zero.
X{[2 2], [1 -2], [-2 -2], [-1 -1]}; Y{0, 1, 0, 1}
Q5. Consider four-class decision problem and train the perceptron network
using perceptron learning rule. Assume initial weights as W11=1, W12=0,
W21=0, W22=1 and biases as b11=1, b21=1.
X1 X2 Y1 Y2
1 1 0 0
1 2 0 0
2 -1 0 1
2 0 0 1
-1 2 1 0
-2 1 1 0
-1 -1 1 1
-2 -2 1 1

MODULE 2
Q1. Write a program in any language for all the questions in Module
1.
Q2. Write a program in any language for solving XOR problem using
multi-layer feed forward neural network concept.