Dr. A.P.J. Abdul Kalam University
Bachelor of Engineering
Eighth Semester Main Examination, June-2021
Soft Computing [IT-802T]
Branch-IT
Time: 3:00 Hrs Max Marks 70
Note: i. Attempt any five questions out of eight.
ii. All question carry equal marks.
iii. Answer should be precise & to be point only.
iv. Assume suitable data if necessary & state them clearly.
Q.1 (a) What is meant by an activation function in an artificial neuron model? Describe the various
activation functions that are employed and compare their merits and demerits.
(b) Explain artificial neural network architecture? With its applications.
Q.2 (a) Explain perceptron network training with and without bias by taking suitable examples.
(b) Explain unsupervised learning mechanism in contrast with a supervised learning mechanism.
Q.3 (a) Compare the similarities and difference between single layer and multilayer perceptrons and also
discuss in what aspects multilayer perceptrons are advantageous over single layer perceptrons.
(b) Discuss the application of neural network in data compression.
Q.4 (a) Explain radial basis function network in brief.
(b) Explain linear separability using an example IS XOR gate linear separable.
Q.5 (a) Explain the architecture of ART with diagram.
(b) Discuss how a neural network may be trained for a pattern recognition task.
Q.6 (a) Describe the self organizing map architecture and explain Kohonen model.
(b) What are crisp relations? How are they different from fuzzy relations? Explain various properties
of crisp relation and fuzzy relation?
Q.7 (a) Explain neuro genetic hybrid and fuzzy genetic hybrid system.
(b) What is fuzzy quantifier? Explain.
Q.8 (a) Explain various types of crossover and mutation techniques.
(b) Explain genetic algorithm in terms of individual, gene, fitness, population, encoding selection,
crossover, mutation.
Scanned Copies:
0 comments:
Pen down your valuable important comments below