CST 444 SOFT COMPUTING

Module I
Sl. No Questions Mark Year
s
KTU
1 Differentiate between the problem solving strategies: Soft 3
(S,FE)
computing and Hard computing.?
2021
2 What is the role of activation function in Artificial Neural 3 KTU(S,FE
Network (ANN)? Write and explain any two activation )2021
functions of ANN.
3 Implement AND function using Mc Culloch-Pitts neuron. Use 6 KTU(S,FE
binary data. )2021
4 Calculate the net input to the neuron Y for the network shown in 5 KTU(S,FE
figure. )
2021

5 Explain the five basic architectures of ANN based on connection. 4 KTU(S,FE

)
2021
6 “The Hebb rule is more suited for Bipolar data than binary data.” 3 DEC 2021
Justify the statement.?

KTU
7 3
Explain the training algorithm of Hebb network. (S,FE)
JAN
2022
8 What are the different types of learning methods employed in 4 KTU
neural networks? (S,FE)
JAN
2022
Implementation of logic gates using i) MP neuron ii) Hebb network MODEL
9 8
--AND,OR,NOR,XOR ,AND, NOT etc. QUESTIO
N
Obtain the output of the neuron for a network with inputs are given as MODEL
10 8
[x1, x2] = [0.7, 0.8] and the weights are [w1, w2] = [0.2, 0.3] with QUESTIO
bias = 0.9. N
Use i) Binary sigmoidal activation function ii) Bipolar sigmoid
activation function

11 Design a Hebb net to implement NOR function using with 6 KTU

bipolar inputs and targets. (R,S)
DEC 2020
7
 A 4-input neuron has weights 1, 2, 3 and 4. The transfer function is
12 6 MODEL
linear with the constant of proportionality being equal to 2. The inputs
QUESTION
are 4, 10, 5 and 20 respectively. Predict the output?

13 Implement AND(x1, x2) where x1, x2 is an element of {0,1} using 4 MODEL

MP neuron. QUESTION
Module II
1 What are the activations used in backpropagation network algorithm 3 KTU (PT)DEC
2021
State the testing algorithm used in perceptron algorithm.
KTU (PT)DEC
2 4
2021
How is error propagated in backpropagation network? Explain the
3 5 KTU (S,Fe)Jan
phase II of BPN training algorithm. 2022

Implement OR function using perceptron training algorithm with

4 4 KTU Dec 2020
binary inputs and bipolar targets. (up to 2 epochs)
Draw the architecture of back propagation network and write
5 training algorithm. 6 KTU Dec 2020

6 Explain the architecture and training algorithm of Adaline network . 8 Model

Use Adaline nerwork to train NOR logic function with bipolar inputs Question
and targets. Perform 2 epochs of training.

7 Discuss the training algorithm and explain the weight updates in back 10 Model
propagation networks Question

Explain how synaptic weights are adapted iteration by iteration using

8 error correction rule in Perceptron convergence algorithm for an OR 10 Model
gate with bipolar inputs and outputs. Initial weights are all zero and Question
learning rate parameter η=0.1

9 Explain Perceptron convergence theorem and discuss Perceptron 4 Model

algorithm based on XOR logic function. Question

10 Implement OR function using Adaline network. Use bipolar inputs. 6 Model

Perform one epoch of training. Question
Module III

8
1 5 KTU DEC
Define (i) core (ii) support (iii) boundary of a fuzzy set. 2020

2 List the properties of fuzzy set. 5 KTU DEC

2020
Differentiate between fuzzy tolerance and equivalence relation. How
3 3 KTU Dec 2020
can we convert fuzzy tolerance relation to fuzzy equivalence relation?
Using your own intuition and your own definitions of the universe of
4 3 KTU Dec 2020
discourse,
plot fuzzy membership functions for “Age of people”:
(i) Very young (ii) Young (iii) Middle-aged (iv) Old (v) Very old

5 Define composition operation in fuzzy relations? Two fuzzy relations 9 KTU (S,FE)
R and S are given below. Obtain the composition T of these fuzzy 2022
relations using

i)Max-min composition
ii)Max-product composition

6 Given two fuzzy sets A and B, compute the following set operations 5 KTU (S,FE)
on A and 2022
B:
a) Algebraic sum b) Algebraic product c) Bounded sum d) Bounded
difference

Module IV
Apply Mamdani fuzzy model to design a controller to determine the Model
1 10
wash time of a domestic washing machine. Assume input is dirt and Question
grease of the cloth. Use three descriptors for input variable and five
descriptors for output variables .Derive the set of rules for controller
action and defuzzification. Design should be supported by figure
wherever possible.

9
2 Using inference approach, find the membership values for each of the 8 Model
triangular shapes (I, R, IR, T) for a triangle with angles 120°, 50°, 10°.
Question
Explain Single-Point Crossover and Two-Point Crossover with Model
3 4
example. question

Differentiate between value encoding and permutation encoding. Model

4 8
Question
Explain the stopping conditions for genetic algorithm.
5 6 Model question
Explain five different types of selection and crossover genetic
6 10 KTU(S,FE)JA
operators.
N2022
With block diagram explain about Fuzzy Inference system.
7 5 KTU
DEC2020
With an example explain Mamdani FIS model.
8 6 KTU(PT)DEC
2021
List out the various encoding techniques in GA with examples for each
9 5 KTU(PT)DEC
2021
MODULE V
1 Explain convex and non convex MOOP? How to find a non 10 Model
dominated set. Question

2 What are the properties of dominance relation? 4 Model

Question
Explain Genetic Neuro-Hybrid System with block diagram. Also write
3 8 Model
the advantages of Genetic- Neuro Hybrid systems
Question
Discuss the classification of Neuro-Fuzzy Hybrid System.
4 6 Model
Question
5 Give the properties of Genetic-neuro hybrid systems with a 7 Model question
block diagram.
With suitable example,explain about different methods of encoding that
6 5 Model
are possible in genetic algorithm.
Question
List the various advantages of Genetic-neuro hybrid systems.
7 3 Model
Question
Explain cooperative Neural Fuzzy systems.
8 5 Model
Question
9 5 Model
Explain a simple genetic algorithm with an example
Question

10