r05410201 - Neural Networks & Fuzzy Logic
Uploaded by
SRINIVASA RAO GANTAr05410201 - Neural Networks & Fuzzy Logic
Uploaded by
SRINIVASA RAO GANTAP.
CODE:37221 R05 SET- 1
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
IV.B.TECH - I SEMESTER REGULAR EXAMINATIONS NOV/DEC, 2009
NEURAL NETWORKS AND FUZZY LOGIC
(Common to EEE, E.CON.E, MEP, AE, ICE, AME)
Time: 3hours Max.Marks:80
Answer any FIVE questions
All questions carry equal marks
---
1. a) Explain the organization of brain in detail.
b) Explain what is an artificial neural network and show how a basic ANN is constructed from
a biological neuron concept. [8+8]
2. a) Briefly discuss about linear separability. Also, suggest a network that can solve EX-OR
problem.
b) Write short notes on artificial neural network architectures. [8+8]
3. a) Explain step by step procedure of single discrete perception training algorithm (SDPTA)
b) Write short notes on “instar” and “outstar” learning. [8+8]
4. State and explain the generalized delta learning rule applied in back propagation algorithm.
[16]
5. a) Explain the working of a Hopfield network with a neat sketch of its architecture.
b) A Hopfield network made up of 5 neurons, which is required to store the following three
fundamental memories.
E1= {+1, +1, +1, +1, +1}T
E2= {+1, -1, -1, +1, -1}T
E3= {-1, +1, -1, +1, +1}T
Evaluate the 5-by-5 synaptic weight matrix of the network. [8+8]
6. a) Distinguish between Crisp logic and Fuzzy logic.
~ ~
b) Consider the fuzzy sets A & B defined on the interval X = [0,5] of real numbers, by the
membership grade functions.
~
~ X
µ A( x) = , µ B( x) = 2− X
X +1
Determine the mathematical formulae and graphs of the membership grade functions of
i) A ∪ B
ii) A ∩ B [8+8]
7. a) Write short notes on fuzzification interface and knowledge base in a fuzzy logic controller.
b) Define defuzzification. Explain any two methods of defuzzification. [8+8]
8. Explain the application of neural networks in character recognition. [16]
P.CODE:37221 R05 SET- 2
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
IV.B.TECH - I SEMESTER REGULAR EXAMINATIONS NOV/DEC, 2009
NEURAL NETWORKS AND FUZZY LOGIC
(Common to EEE, E.CON.E, MEP, AE, ICE, AME)
Time: 3hours Max.Marks:80
Answer any FIVE questions
All questions carry equal marks
---
1. a) What is an artificial neural network? Describe the characteristics of artificial neural
networks?
b) Explain the architecture of Hodgkin-Huxley neuron model. [8+8]
2. a) Compare activation dynamics and synaptic dynamics models.
b) Classify the learning methods. Give a brief explanation about each. [8+8]
3. a) Describe a perceptron neuron and give its limitations.
b) Explain step by step procedure of single continuous perceptron training algorithm
(SCPTA). [8+8]
4. With suitable diagram explain the concept of back propagation? Derive update equations for
weight elements of multi-layer feed forward neural network. Discuss its applicability for the
problems of pattern recognition. [16]
5. a) Explain the general concept of associate memory. Define associate matrix and associate
rules.
b) What are the modes of operation of a Hopfield network? Explain the algorithm for storage
of information in a Hopfield network. [8+8]
6. a) Explain the following terms in sets.
i) CON
ii) DIL
iii) Membership function
iv) CRISP.
b) Explain union, intersection and complement with reference to fuzzy set and classical sets.
[8+8]
7. a) Compare and contrast fuzzy logic control and classical control system.
b) Explain min-max method of implication with a suitable example. [8+8]
8. Design and develop pressure process control by fuzzy logic control model. Formulate
necessary membership functions and required fuzzy rules for the application. [16]
P.CODE:37221 R05 SET- 3
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
IV.B.TECH - I SEMESTER REGULAR EXAMINATIONS NOV/DEC, 2009
NEURAL NETWORKS AND FUZZY LOGIC
(Common to EEE, E.CON.E, MEP, AE, ICE, AME)
Time: 3hours Max.Marks:80
Answer any FIVE questions
All questions carry equal marks
---
1. a) Explain the architecture of integrate and fire neuron model.
b) Briefly explain the historical developments of ANN, with a mention of their potential
applications. [8+8]
2. a) What is an activation function. Mention various activation functions with their merits and
demerits.
b) Describe the Rosenblatt’s perceptron model of an artificial neuron. [8+8]
3. a) Distinguish between linearly separable and linearly non-separable problem. Give examples
for each.
b) Write perceptron training algorithm for several output classes. [8+8]
4. What is back propagation? With a schematic two-layer feed forward neural network, derive its
learning algorithm. Also discuss its learning difficulties and improvements.
[16]
5. What are the self organizing maps? Explain the architecture and the training algorithm used
for Kohonen’s SOMs. [16]
6. a) Explain the properties of commutativity, associativity, distrubutivity, idempotence and
identity with respect to crisp sets and fuzzy sets.
b) Define membership function and universe of discourse in fuzzy logic. [8+8]
7. a) With a neat sketch discuss the major components of fuzzy controller.
b) What is the role of membership function in fuzzy logic? Mention various membership
functions. [8+8]
8. Design a fuzzy controller for a temperature control system of a room. Assume your own
control actions due to which the temperature of the room may vary. Design in fuzzy rule-based
system to keep the room at a comfortable temperature. [16]
P.CODE:37221 R05 SET- 4
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
IV.B.TECH - I SEMESTER REGULAR EXAMINATIONS NOV/DEC, 2009
NEURAL NETWORKS AND FUZZY LOGIC
(Common to EEE, E.CON.E, MEP, AE, ICE, AME)
Time: 3hours Max.Marks:80
Answer any FIVE questions
All questions carry equal marks
---
1. a) Compare the biological neural network with artificial neural network.
b) Explain the architecture of spiking-neuron model. [8+8]
2. a) Describe the Mc-Culloch – pitt’s model of neuron. Design a network using this model to
realize the NAND gate.
b) Classify the learning methods. Give a brief explanation about each. [8+8]
3. State and prove perceptron convergence theorem. [16]
4. Define Kolmogrov theorem and also the difficulties associated with learning in multilayer
perceptrons. [16]
5. a) State and prove bi-directional associative memory energy theorem.
b) With suitable diagram, explain the learning of Boltzmann’s machine. [8+8]
6. a) Explain basic fuzzy set operations.
b) Let X={1, 2, 3, ………10}. Determine the cardinalities and relative cardinalities of the
following fuzzy sets.
~
i) A = {(2, 0.4), (3, 0.6), (4, 0.8), (5, 1.0), (6, 0.8), (7, 0.6), (8, 0.4)}
~
ii) B = {(2, 0.4), (4, 0.8), (5, 1.0), (7, 0.6)} [8+8]
7. List the various defuzzification techniques. Explain each of them in detail. [16]
8. Design and develop an air conditioner controller by fuzzy logic control model. Formulate
necessary membership functions and required fuzzy rules for the application.
[16]
