UEC823
Uploaded by
vivesod153UEC823
Uploaded by
vivesod153Roll Number:
Thapar University, Patiala
Department of Electronics and Communication Engineering
AUXILIARY EXAMINATION, FEB-2023
B. E (Final Year)/: Semester-7th (ECE) Course Code: UEC-823
Course Name: Soft Computing
3rd March,2022 Friday, 1730 Hrs
Time:3 Hours, M. Marks: 100 Name of Faculty: GK
8Note: Attempt all questions in SERIAL ORDER. Marks are mentioned against each
question. Assume missing data, if any, suitably.
1 What is soft computing? Compare ANN with Biological neuron network. (10)
2 a. Discuss different activation functions in ANN. (6)
B For 'AND' gate develop the perceptron model using bipolar inputs and outputs. (4)
.1 + 0. 5 + _ + 0.2
3. Using fuzzy operations for sets A=[° 1+ °
3 5 66 18 9 ] and
0.15 0.4 0.65 0.7
B= [-+ - + 6 + 8 ]
3 5
(i)Find AnB, (ii) Find AuB (iii) A-B .
Also find cardinality and power set for set A (10)
4. Explain Centroid and weighted average defuzzification methods. (10)
4. What is backpropagation algorithm? Explain. Also mention atleast 2 limitations. (10)
5 What are the different Selection mechanisms used for GA? Explain Roulettee wheel
selection with example. (10)
6. What is travelling salesman problem? Explain the following with example for TSP:
a) Partially Matched crossover
b)Overlapping crossover (10)
7. Draw the comparison of fuzzy and Crisp sets. What is Mamadani Fuzzy Inference
system? (10)
8. What are KSOMs? Assume a KSOM with 3 cluster units and 4 input units. The weight matrix
is [0.9 0.4 0.5;0.7 0.3 0.8; 0.6 0.5 0.4; 0.4 0.6 0.7 ]. Assuming Learning rate=0.2 and
Euclidean distance as the discriminant function. Perform one epoch of learning and find
updated weights, when input applied is [0.4 0.2 0.1 0] (10)
9a) Differentiate between supervised and unsupervised learning. (5)
b) State and Explain the Hebbs learning law. (5)
10. . Assume a feed-forward neural net with 4 input neurons and one output neuron.
Assume the input values of 0.4, 0.5 , 0.8 and 0.9 with weight values of 0.3, -0.1, 0.2
and -0.2 respectively. If bias value is 0.35 compute the output for bipolar sigmoidal
activation function. (10)
